An Improved Modular Neural Network Model for Adaptive Trajectory Tracking Control of Robot Manipulators

2009 ◽  
pp. 1063-1070 ◽  
Author(s):  
Dimitrios Alexios Karras
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Trajectory Tracking ◽  
Tracking Control ◽  
Robot Manipulators ◽  
Trajectory Tracking Control ◽  
Modular Neural Network

Trajectory tracking control of robot manipulators using a neural-network-based torque compensator

1998 ◽  
Vol 212 (5) ◽  
pp. 361-372 ◽  
Author(s):  
Q Li ◽  
S K Tso ◽  
W J Zhang
Keyword(s):  
Neural Network ◽  
Trajectory Tracking ◽  
Tracking Control ◽  
High Performance ◽  
Robot Manipulator ◽  
Robot Manipulators ◽  
Lyapunov Stability Theory ◽  
System Stability ◽  
Trajectory Tracking Control ◽  
Modelling Uncertainties

In this paper, an adaptive neural-network-based torque compensator is developed for the trajectory-tracking control of robot manipulators. The overall control structure employs a classical non-linear decoupling controller for actuating torque computation based on an approximated robot dynamic model. To suppress the effects of uncertainties associated with the estimated model, a supplementary neural network algorithm is developed to generate compensation torques. The weight adaptation rule for this neuro-compensator is derived on the basis of the Lyapunov stability theory. Both global system stability and the error convergence can then be guaranteed. Simulation studies on a two-link robot manipulator demonstrate that high performance of the proposed control algorithm could be achieved under severe modelling uncertainties.

Sign in / Sign up

Export Citation Format

Share Document